Live
Cloudflare Images Analytics Dashboard Adds 30‑Day Delivery VisibilityAmazon ECS Adds Native Blue/Green, Canary, and Linear Deployments for Safer Container RolloutsGemini agentic tools add LLM‑driven automation to Partner Marketing StudioGitHub Actions makes RHEL 9/10 runner images generally available for larger workloadsIntegrating Decider Models to Reduce Latency and Cost in AI Agent WorkflowsAWS Emulator Enables Local Testing of 200+ Services, but Lacks IAM ValidationSageMaker AI monitoring adds automatic latency drift, idle GPU, and configuration gap detection for inference endpointsEmbedding Proactive Governance into Enterprise‑Scale Agentic AICloudflare Images Analytics Dashboard Adds 30‑Day Delivery VisibilityAmazon ECS Adds Native Blue/Green, Canary, and Linear Deployments for Safer Container RolloutsGemini agentic tools add LLM‑driven automation to Partner Marketing StudioGitHub Actions makes RHEL 9/10 runner images generally available for larger workloadsIntegrating Decider Models to Reduce Latency and Cost in AI Agent WorkflowsAWS Emulator Enables Local Testing of 200+ Services, but Lacks IAM ValidationSageMaker AI monitoring adds automatic latency drift, idle GPU, and configuration gap detection for inference endpointsEmbedding Proactive Governance into Enterprise‑Scale Agentic AI
AI Engineering

Anthropic Enterprise Gateway for AWS and GCP

AI SummaryPowered by AI

The new Anthropic self-hosted gateway streamlines the deployment of Claude Code on major cloud platforms by centralizing identity management. This solution eliminates credential sprawl, offering a robust architecture that aligns with advanced security practices required in modern DevOps environments.

Deploying AI coding assistants like Claude Code across large engineering organizations often introduces significant operational friction without proper architectural planning. Historically, scaling these tools requires managing individual cloud credentials for every developer and manually pushing configuration settings to each workstation. This approach creates a security surface area that is difficult to audit or control effectively.

To address this challenge, Anthropic has released an enterprise gateway designed specifically for deployment on Amazon Bedrock and Google Cloud infrastructure. The Anthropic Enterprise Gateway functions as a single stateless container backed by PostgreSQL, acting as the central point of contact between developers and model providers. By sitting in front of these services, it handles identity verification, policy enforcement, usage tracking, and spend management automatically.

Centralized Identity Management with OIDC

The gateway fundamentally changes how access is granted to AI models by acting as an OpenID Connect relying party. Instead of distributing long-lived API keys or secrets that must be rotated manually when a developer leaves the company, this system issues short-lived sessions based on standard identity providers like Google Workspace, Microsoft Entra ID (formerly Azure AD), Okta, and any other standards-compliant OIDC provider.

From an architectural perspective, replacing static credentials with dynamic session tokens significantly reduces risk. If a token is compromised or leaked in logs, it expires quickly rather than granting persistent access indefinitely.

Simplified Infrastructure Deployment


The deployment model for the Claude Code Gateway prioritizes simplicity and statelessness to ensure high availability across Kubernetes clusters or container orchestration environments. Organizations can deploy this single component on their own infrastructure, which then connects directly to existing AI models hosted in AWS Bedrock or GCP Vertex AI.

This design pattern is particularly relevant for professionals preparing for Kubernetes certifications, as it demonstrates how stateless sidecars manage complex backend interactions. The PostgreSQL database component handles the necessary session storage and audit logs, ensuring that every interaction with an LLM model can be traced back to a specific user identity.

Operational Visibility for Finance Teams


Enterprises often struggle to track AI spending because usage data is fragmented across different developer laptops. The gateway aggregates this information into unified dashboards, providing finance teams and security operations centers with real-time visibility into who is using the models and how much they are consuming.

This capability transforms cost management from a reactive process of guessing bills at month-end to an active governance model where budgets can be enforced via policy rules defined within the gateway itself.

What This Means For You


For DevOps engineers, this release represents a shift toward more secure and manageable AI tooling. By adopting patterns like short-lived sessions through OIDC providers, teams align their practices with industry best standards for zero-trust architecture. The Claude Code Gateway effectively removes the manual overhead of provisioning credentials per developer while maintaining strict control over access policies.

Originally published atDEVOPS